Press your lips together. Release the air with voice. You produce b. Now do the same thing but let the air come through your nose with the lips closed. You produce m. The two sounds share the same articulation site (lips) and the same articulation gesture (closure + release), differing only in whether the air exits through the mouth (b) or the nose (m). The two are phonetically adjacent in a way that b and, say, k are not. This adjacency is part of what makes the M-B interchange a recurring pattern in many languages.
In Akan, the interchange is well-attested:
The pattern is uncontested in Akan lexicography. The B-form and the M-form are recognized variants of the same lexeme, often distinguishing singular from plural, often dialectal variants of each other.
Odwirafo argues that the same M-B interchange visible within Akan is also visible in ancient Egyptian, and that this systematic-phonological parallel is evidence of shared phonological structure between the two languages — therefore evidence of shared genetic descent.4
The key Egyptian case he cites: Mart (or Mert) is written in Egyptian medutu with the M-R-T root, meaning "the divine eyes" / "sun and moon." In the Coptic dialect (the last spoken form of Egyptian, ca. 2,000 years ago), the same word appears as Bart. The M-B interchange in Mart ↔ Bart is the same phonological alternation visible in Akan Moja ↔ Boja.5
Odwirafo's methodological argument: when an Egyptologist encounters a word spelled with the M-medut in one inscription and the B-medut (the leg-with-foot symbol) in another inscription, the Egyptologist recognizes this as variant-spelling of the same word. The Akan-language M-B interchange is the same operation, attested live across thousands of years of independent transmission. The shared phonological pattern, in this reading, is evidence the two languages descend from a common source.
[TRADITION INTERPRETATION — diverges from mainstream scholarship]: Mainstream historical linguistics does not accept shared phonological-process patterns as evidence of shared genetic descent in the absence of systematic sound correspondences across cognate sets. The M-B interchange is a common cross-linguistic phonological phenomenon — it appears in many unrelated languages because the phonetic adjacency of m and b makes the alternation easy in any language with both phonemes. Latin Septembrem ↔ Modern Greek Septemvris, the various m/b alternations in Indo-European nasal-and-stop interactions, and parallel alternations in unrelated language families all show similar patterns. Shared phonological-process patterns are weak evidence for shared descent specifically because the same patterns can arise independently in unrelated languages.
The within-Akan M-B interchange is uncontested. The within-Egyptian (specifically Coptic-vs-medutu spelling variation) M-B interchange is uncontested. The cross-language argument that these two within-language patterns evidence shared genetic descent is the Cat 1 quarantine. See Odwirafo Source Tensions Category 1.
The structural argument has three steps:
Step 1: Within Akan, the M-B interchange follows recognizable grammatical and dialectal patterns (plural-formation, dialect-variation). It is not random; it is structurally embedded in the language.
Step 2: Within Egyptian, the M-B interchange is visible at the spelling-medutu level (M-medut and B-medut as variant spellings of the same lexeme) and at the dialectal-historical level (Coptic Bart for earlier Mart).
Step 3: The structural similarity of these two within-language patterns — the same phonological alternation operating in regular ways — is evidence that the two languages share the phonological architecture that allows the alternation. If they share the phonological architecture, they likely share the genetic source.
The counter-argument is that Step 3 does not follow. The same phonological alternation can arise independently in any language with both m and b phonemes because of the underlying phonetic adjacency. Structural similarity at the phonological-process level is generic; it is not lineage-specific.
The M-B interchange page serves as the dedicated treatment of the largest systematic phonological-pattern Odwirafo identifies. It feeds:
The strongest cross-language case in Odwirafo's pattern is the Mart ↔ Bart alternation between medutu-Egyptian and Coptic. In the medutu (hieroglyphic) period, the word for "the divine eyes" / "sun and moon" is spelled with the M-medut: M-R-T, vocalized as Mart or Mert. In Coptic — the late stage of the language, ca. 2,000 years ago — the same word appears as Bart. The phonological shift from M to B (or vice versa) is documented within the historical record of one language.
Walk through what this case tells us. Within Egyptian, the M-B alternation is not just a spelling variation; it is a phonological shift that occurred at some point in the language's history. This is mainstream-attested Coptic linguistics — Coptic phonology shifted in various ways from earlier Egyptian, and the M-B alternation in Mart/Bart is one of those documented shifts.
Odwirafo's argument: if Egyptian itself contains an internal M-B alternation in its historical record, and Akan contains the same M-B alternation as a live feature, then the two languages share a phonological feature. If they share a phonological feature, the shared-source hypothesis is strengthened.
The counter-argument: the M-B shift in Egyptian is a documented one-time historical change within one language's development. The Akan M-B alternation is a live productive feature of a different language. Two languages having both undergone or both featuring an M-B alternation does not require shared descent — it requires only that both languages have the phonological raw materials (m and b phonemes adjacent in articulation space) to make the alternation possible. Both kinds of pattern have arisen independently in many unrelated languages.
The case study is methodologically clean precisely because it isolates the structural argument. The dispute is not about facts (both alternations are documented); it is about what the parallel between them implies. Reasonable methodological positions exist on both sides.
A vault-engaged practitioner or scholar working with the broader Akan-Kemetic catalog will encounter many pairs where the M-B interchange is the bridging phonological rule (e.g., a pair where the Akan word starts with M and the Egyptian word starts with B, or vice versa, and the equivalence depends on accepting that M-B substitution is licensed). This page is the reference for what is being assumed when those pairs are deployed.
Three operational uses:
Use 1 — Audit cross-language pairs. When the broader catalog presents a pair where the Akan and Egyptian forms differ at the M-B alternation, the audit question is: is the alternation licensed by within-Akan and within-Egyptian patterns, or only by the cross-language theory? If only the cross-language theory licenses it, the pair is weaker evidence.
Use 2 — Separate within-language facts from cross-language theory. The within-Akan M-B interchange (Boja ↔ Moja) is a fact of the Akan language regardless of any cross-language argument. Use the within-Akan fact for within-Akan work (e.g., the Abosom / Mu / Musu etymological cluster within Akan cosmology); use the cross-language extension only when the cross-language framework is being explicitly engaged.
Use 3 — Engage the methodological dispute consciously. The M-B interchange is the cleanest test case in Odwirafo's broader catalog for the methodological dispute about whether shared phonological patterns evidence shared genetic descent. A serious engagement with the catalog is partly a serious engagement with this methodological question.
Teaching-level failure: A teacher presents the M-B interchange as straightforward evidence of Akan-Egyptian shared descent without acknowledging the mainstream-linguistics counter-argument. The pattern becomes a tribal-marker rather than a methodologically-disputable claim. The corrective is to teach the pattern with the methodological dispute present, so students learn what kind of argument is being made.
Practice-level failure: A practitioner uses the M-B interchange to generate ad-hoc Akan-Egyptian pair-equivalences for any pair where the M and B happen to substitute. Without methodological discipline, the framework over-generates — every M-word in one language can be paired with every B-word in the other language, regardless of semantic constraint. The corrective is the per-pair semantic-precision discipline: phonological alternation alone is insufficient evidence; semantic precision must also be present.
Translation-level failure: A scholar dismisses the M-B interchange material wholesale because the cross-language extension is rejected. The within-Akan and within-Egyptian patterns are well-attested facts about each language; engaging them as language-internal phenomena does not require accepting the cross-language extension. The corrective is the layered engagement: accept the within-language facts; engage the cross-language argument as a separate methodological question.
The within-Akan M-B interchange is mainstream-attested (Christaller 1881; subsequent Akan lexicography; documented in dialectal and grammatical descriptions of Akan). The within-Egyptian M-B alternation (specifically Mart ↔ Bart across medutu and Coptic) is mainstream-attested in Egyptology and Coptic linguistics. Both within-language patterns are uncontested.
[TRADITION INTERPRETATION — diverges from mainstream scholarship]: The cross-language argument that the parallel constitutes evidence of shared genetic descent is the Odwirafo-lineage interpretive position. See Odwirafo Source Tensions Category 1.
Open question (filed to META): How does mainstream historical linguistics treat the systematic M-B interchange claim as evidence for genetic relationship between Niger-Congo (Akan) and Afro-Asiatic (Ancient Egyptian)? Are there comparable shared-phonological-process patterns between language-families that mainstream linguistics has accepted as evidence of shared descent, and if so, what additional methodological criteria did the acceptance require?
Single source for the cross-language argument (Odwirafo, Akanfo Nanasom W2). The implicit interlocutors are mainstream historical linguistics (Hock; Campbell; the broader Comparative Method tradition) and mainstream Akan lexicography (Christaller; Bartels). Convergence on the within-language facts of each pattern; divergence on the cross-language genetic-linguistic interpretation.
The pattern shared phonological-process patterns as evidence vs. as coincidence is a structural question that recurs across linguistics and historical research with revealing parallels.
anthropology: Methodological Disputes in Comparative Fieldwork — the broader pattern that comparative-research methodologies can generate datasets that observers from different methodological commitments will interpret radically differently. The Akan-Kemetic M-B interchange case is structurally the same as disputes over comparable patterns in comparative-religion (parallel mythological motifs as evidence of shared origin vs. as evidence of universal cognitive substrate), comparative-mythology (parallel narrative structures as diffusion-evidence vs. as universal-form-evidence), and comparative-art-history (parallel iconographic conventions). The handshake reveals: comparative-research disputes tend to come down to methodological commitments about what counts as evidence, not to disputes about facts. The insight neither domain produces alone: a scholar engaging any comparative-research dispute productively needs to be conscious of which methodological frame is being deployed and what evidence-standards it permits; defaulting to the institutionally-dominant frame without conscious choice often misrepresents the underlying dispute.
history: Historiographical Burden-of-Proof Debates — the historiographical pattern that burden-of-proof allocations in comparative-historical claims (who has to prove the alternative explanation? what is the null hypothesis?) determine which kinds of evidence are required for which conclusions. The Akan-Kemetic M-B interchange depends on burden-of-proof allocation: under the mainstream-linguistics burden-of-proof, the burden is on the claimant to demonstrate shared descent through systematic sound correspondences across controlled cognate sets; under Odwirafo's framework, the burden is differently allocated. The handshake reveals: burden-of-proof allocations in comparative-historical claims are themselves methodological commitments that can be argued for or against, not given facts of scholarship. The insight: where two researchers disagree about whether a piece of evidence is sufficient, the disagreement often runs through a deeper disagreement about which framework gets to set the burden-of-proof allocation.
The Sharpest Implication
If a shared phonological-process pattern between two languages can be either strong evidence of shared descent or generic universal-pattern coincidence, depending entirely on which methodological framework is brought to it, then the Akan-Kemetic comparative-linguistic dispute is not (only) a dispute about Akan and Egyptian. It is a clean instance of a deeper dispute about what comparative-evidence can support in principle. The challenge for any reader is to recognize that picking a side in the Akan-Kemetic dispute is also picking a side in the broader methodological dispute about comparative-research evidence-standards. The choice is consequential well beyond the specific Akan-Egyptian case.
Generative Questions